README.md

Classification

Perform classification on the fully dimensional data. Binary classification on bot label, botometer labels, and opinion score. Multiclass classification on fourway labels (opinion x bot status).

Each job will generate a CSV file of predictions. Which we are uploading to Google Drive under "text-embedding-ada-002/results/classifications/TARGET_NAME/MODEL_NAME/predictions.csv" for further analysis of any mis-classifications / confusions.

Logistic Regression

python -m app.classification.logistic_regression

FIG_SHOW=true FIG_SAVE=true python -m app.classification.logistic_regression

Decision Tree

python -m app.classification.decision_tree

Random Forest

python -m app.classification.random_forest
FIG_SHOW=false FIG_SAVE=false python -m app.classification.random_forest

XGBoost

python -m app.classification.xgboost

FIG_SHOW=false FIG_SAVE=false python -m app.classification.xgboost